aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2016-11-01Update formatting of files before committing changes for autotest. No ↵Manoj Gupta
functional change. BUG: chromium:647429 TEST: crosperf unit tests Change-Id: I827f9ec4d00a0d8c1ae8a9930d7832289e694dc8 Reviewed-on: https://chrome-internal-review.googlesource.com/301457 Commit-Ready: Manoj Gupta <manojgupta@google.com> Tested-by: Manoj Gupta <manojgupta@google.com> Reviewed-by: Luis Lozano <llozano@chromium.org>
2016-10-28[crosperf] Remove stumpy machines.Caroline Tice
We are decommissioning our stumpy lab machines and returning them to the ChromeStop. This CL removes the stumpy machines from the default_remotes file. BUG=chromium:660536 TEST=None Change-Id: I5ecf1536fa399ff70a89271394c8d4e6e09e64e1 Reviewed-on: https://chrome-internal-review.googlesource.com/301258 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Yunlian Jiang <yunlian@google.com>
2016-10-27[crosperf] Fix buildbot_utils for rotating tester waterfall.Caroline Tice
The rotating testers trybot waterfall location has moved. This CL updates the script to look for their build logs in the correct builder location. It also makes sure we don't crash/die because their build status is non-zero (ie. a HWTest failed). This also fixes a tiny bug in the results reporting. BUG=chromium:660085 TEST=Tested in the role account. Change-Id: I396ec0ad588e54e88bc44dcc0e72eda45d077173 Reviewed-on: https://chrome-internal-review.googlesource.com/300598 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Yunlian Jiang <yunlian@google.com>
2016-10-27[toolchain-utils] Update page_cycler in nightly tests.Caroline Tice
Currently page_cycler_v2.typical_25 tries to run too many times, causing some of the nightly builders to take nearly 24 hours to complete (they take about 6 hours without page_cycler). Also, page_cycler_v2.typical_25 cannot run locally on some boards (such as daisy). This CL removes page_cycler_v2.typical_25 from the all_toolchain_perf set, and adds it separately to the nightly tests, with only 2 iterations, no retries, and forcing it to not run locallys on the DUT. BUG=chromium:660087 TEST=Tested the new page cycler settings in a crosperf run. Change-Id: Ia68abce4e8ee29a671e58db13f269e531d31736e Reviewed-on: https://chrome-internal-review.googlesource.com/300855 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Yunlian Jiang <yunlian@google.com>
2016-10-24Handle exceptions in crosperf.Manoj Gupta
Change error message in buildbot. Catch exceptions in crosperf exceution to do a graceful exit. BUG=chromium:622875 TEST=ran run_tests Change-Id: I03093ab27edcc85e318ad43097ab8cd316792dcb Reviewed-on: https://chrome-internal-review.googlesource.com/299275 Commit-Ready: Manoj Gupta <manojgupta@google.com> Tested-by: Manoj Gupta <manojgupta@google.com> Reviewed-by: Caroline Tice <cmtice@google.com> Reviewed-by: Luis Lozano <llozano@chromium.org>
2016-10-18[toolchain-utils] Fix rotating builders to work with no patches.Caroline Tice
Currently if no patch is passed to cbuildbot, it hangs; similarly if no patch is given to the rotating builder script, it dies. This CL fixes both of those. It also temporarily stops the rotating builders from sending emails until we can figure out how to fix their reports. BUG=chromium:657060 TEST=Tested changes in role account;seemed to work. Change-Id: I4ba70b562f9cb44d396361cd6461230d7080c6e9 Reviewed-on: https://chrome-internal-review.googlesource.com/297584 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Caroline Tice <cmtice@google.com> Reviewed-by: Ting-Yuan Huang <laszio@google.com>
2016-10-12[toolchain-utils] New script to verify which compiler was used.Caroline Tice
This script takes either a single file or a directory tree and verifies that all the compilation units in the file(s) were built with the specified compiler. BUG=chromium:652835 TEST=Tested on subtree within a build tree. Change-Id: Ib365b9c4590b539f5f515df24f6214846183d9e7 Reviewed-on: https://chrome-internal-review.googlesource.com/293115 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Ting-Yuan Huang <laszio@google.com>
2016-10-12crosperf: update lab machine names.Ting-Yuan Huang
Two elm are added. TEST=none BUG=chromium:655227 Change-Id: I2475aeb0e7aa4964b93babfb6fb626a2de7f5dcb Reviewed-on: https://chrome-internal-review.googlesource.com/296135 Commit-Ready: Ting-Yuan Huang <laszio@google.com> Tested-by: Ting-Yuan Huang <laszio@google.com> Reviewed-by: Caroline Tice <cmtice@google.com>
2016-10-10[toolchain-utils] Add --hwtest flag to rotating tester builds.Caroline Tice
The rotating testers need to add --hwtest to the trybot builds to get the hwtests to actually run. This CL adds that flag. BUG=None TEST=Tested in role account. Change-Id: I6ef93f529b6616dd8d1c357743593ed369d1022a Reviewed-on: https://chrome-internal-review.googlesource.com/295075 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Ting-Yuan Huang <laszio@google.com>
2016-10-06[toolchain-utils] Update rotating testers to use new configs.Caroline Tice
This CL updates the rotating testers to use the new chromes_configs for building trybots & running HW tests. BUG=chromium:653296 TEST=Ran scripts in my directory; it worked. Change-Id: Ia60af8b921a966c86b335ef51db5372bf740ed95 Reviewed-on: https://chrome-internal-review.googlesource.com/293775 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Yunlian Jiang <yunlian@google.com>
2016-10-05[crosperf] Fix xbuddy to work with new -toolchain builders.Caroline Tice
The xbuddy translating code was improperly prepending to board name to the waterfall builder trybot images (the ones with the generic architecture names). This CL fixes that. BUG=chromium:653210 TEST=Tested crosperf with 'build: x86-llvm-toolchain/R55-8865.0.0-rc1'; failed without this CL, worked with it. Change-Id: I706c9d882f9c99f8af11eb1782ea6edd1eedbfe5 Reviewed-on: https://chrome-internal-review.googlesource.com/293815 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Yunlian Jiang <yunlian@google.com>
2016-10-05crosperf: fix the error unexpected keyword argument 'command'Yunlian Jiang
BUG=none TEST=the error is gone. Change-Id: I8dad69d2b45ed821be0c7be77e07b33cf0663eb9 Reviewed-on: https://chrome-internal-review.googlesource.com/293515 Reviewed-by: Yunlian Jiang <yunlian@google.com> Commit-Queue: Yunlian Jiang <yunlian@google.com> Tested-by: Yunlian Jiang <yunlian@google.com>
2016-10-04crosperf: Decrease 10 seconds delay for pagecyclers.Yunlian Jiang
New telemetry adds 10 seconds delay after rendering a page. We do not need that so this CL changes the 10 secnods delay to 2 seconds. BUG=None TEST=the 10 seconds delay is changed to 2 seconds.. Change-Id: I5c835acdf4068fba1aceb0d2a5a45e7113a558d3 Reviewed-on: https://chrome-internal-review.googlesource.com/291818 Reviewed-by: Ting-Yuan Huang <laszio@google.com> Commit-Queue: Yunlian Jiang <yunlian@google.com> Tested-by: Yunlian Jiang <yunlian@google.com>
2016-09-28[bisection tool] Don't make failure to flash bad image fatal error.Caroline Tice
Currently during initial setup, if the known 'bad' image fails to flash onto the device, the user is given the option to 'skip'. This causes the entire bisection to fail/stop. This CL changes the skip to allow bisection to continue (it is not unreasonable for a bad image to fail to flash). BUG=chromium:641539 TEST=None Change-Id: Iddc5312ecf593192d563766d677e8d401824ae7d Reviewed-on: https://chrome-internal-review.googlesource.com/290617 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Yunlian Jiang <yunlian@google.com> Reviewed-by: Ting-Yuan Huang <laszio@google.com>
2016-09-26[crosperf] Add new results fields for other page_cycler_v2 tests.Caroline Tice
This Cl adds the default result fields for the other page_cycler_v2 tests (already done for page_cycler_v2.typical_25). It also temporarily removes some tests from the all_pagecyclers set, as they appear to have been removed from telemetry. Will remove these permanently when this is confirmed. BUG=chromium:650307 TEST=Ran crosperf with all_pagecyclers. Change-Id: I9e6c3ae140f66a11d792350b74c8897df74d944c Reviewed-on: https://chrome-internal-review.googlesource.com/289865 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Yunlian Jiang <yunlian@google.com>
2016-09-23[crosperf] Make generate_report report actual failures.George Burgess IV
We would never increment `num_success`, so we'd always return failure. This passed tests, since there were no tests that checked for success. Also, testRunActionsRunsAllActionsRegardlessOfExceptions wasn't really doing what it was supposed to; exceptions wouldn't get raised (since the mocked WriteFile turns into a nop), so all of the WriteFile actions were "passing". Because of the aforementioned bug, this test seemed to work, as well. BUG=None TEST=generate_report_unittest.py passes; cros lint is happy. Change-Id: If4d71af3415d97be8d5ac3bec796fe4589d50aaa Reviewed-on: https://chrome-internal-review.googlesource.com/289802 Commit-Ready: George Burgess <gbiv@google.com> Tested-by: George Burgess <gbiv@google.com> Reviewed-by: Caroline Tice <cmtice@google.com>
2016-09-23[toolchain-utils] Add licensing information.Caroline Tice
BUG=b:31321592 TEST=None Change-Id: I9d6b407f389c7cb42007cb7ded59b8261fb70b49 Reviewed-on: https://chrome-internal-review.googlesource.com/287475 Commit-Ready: Luis Lozano <llozano@chromium.org> Tested-by: Luis Lozano <llozano@chromium.org> Reviewed-by: Luis Lozano <llozano@chromium.org>
2016-09-22rotation_tester: use toolchain-tests instead of paygen_au_canaryYunlian Jiang
BUG=None TEST=it uses toolchain-tests now. Change-Id: I5b0e350e0fc0b365be5e65587b62e022f342338e Reviewed-on: https://chrome-internal-review.googlesource.com/289837 Commit-Ready: Yunlian Jiang <yunlian@google.com> Tested-by: Yunlian Jiang <yunlian@google.com> Reviewed-by: Caroline Tice <cmtice@google.com>
2016-09-22Revert "[crosperf] Remove --autotest_dir arg from non-telemetry tests."Caroline Tice
This reverts commit 70462462d54070a2440fccb506dcea2ea076cc05. Upon further discussion I realize this is the wrong fix for this problem. Change-Id: I7e9c94e763378e4834858acd826985c55bfcea4d Reviewed-on: https://chrome-internal-review.googlesource.com/289636 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Yunlian Jiang <yunlian@google.com>
2016-09-22cros-utils: fix an out of bound error.Yunlian Jiang
The input rgb2short can reach to 255. When the value exceeds 250, it causes out of bound error. This CL fixes that. BUG=chromium:649374 TEST=The out of bound error is fixed. Change-Id: Iee31337e82907e1803a84b5837542733bb224e8f Reviewed-on: https://chrome-internal-review.googlesource.com/289536 Commit-Ready: Yunlian Jiang <yunlian@google.com> Tested-by: Yunlian Jiang <yunlian@google.com> Reviewed-by: Caroline Tice <cmtice@google.com>
2016-09-21[crosperf] Update default values for page_cycler test.Caroline Tice
page_cycler_v2.typical_25 has different field names than page_cycler v1. This updates the default field names used by crosperf to generate reports, to use the new field names for page_cycer. BUG=None TEST=Tested in my source tree. Change-Id: Id91fa7ad7d2ebc915ccad054bb2408a22629272a Reviewed-on: https://chrome-internal-review.googlesource.com/289235 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Yunlian Jiang <yunlian@google.com>
2016-09-21toolchain-utils: remove colortrans.pyYunlian Jiang
This removes colortrans.py from cros_utils. We add a single function inside misc.py instead. BUG=chromium:647308 TEST=crosperf still works, the color text is shown on terminal. Change-Id: I1871bd2c57608f59ded6a8cf720c52cb2c07c3ff Reviewed-on: https://chrome-internal-review.googlesource.com/287538 Commit-Ready: Yunlian Jiang <yunlian@google.com> Tested-by: Yunlian Jiang <yunlian@google.com> Reviewed-by: Luis Lozano <llozano@chromium.org>
2016-09-16[crosperf] Remove --autotest_dir arg from non-telemetry tests.Caroline Tice
Some autotests that are not telemetry tests will fail if we pass the --autotest_dir arg. This CL removes that arg from non-telemetry tests. BUG=chromium:647429 TEST=ran graphics_WebGLAquarium ok. Change-Id: I00c79a6dc8cc4642b5b9dd342a92df91d0cb9c2c Reviewed-on: https://chrome-internal-review.googlesource.com/287597 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Yunlian Jiang <yunlian@google.com>
2016-09-15[crosperf] Don't block on local server if not using local machines.Caroline Tice
Occasionally the local AFE server goes down; this currently causes the afe_lock_machine.py script to fail, whether the user is trying to lock a local machine or not. This CL fixes that, so if the user specified a HW Lab machine, we don't try to look at the local server at all. Also cleans up some formatting. BUG=chromium:647427 TEST=Tested getting status of lab machine when local server was down. Change-Id: Iff7799b2deea142bcbd85f0e3373cf0462590b72 Reviewed-on: https://chrome-internal-review.googlesource.com/287456 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Luis Lozano <llozano@chromium.org>
2016-09-14[bisection] Fix various minor bugs in Android bisection.Caroline Tice
Recent testing has found a few minor bugs in the Android bisection tool. This CL fixes those bugs. BUG=chromium:646948 TEST=Ran full Android bisection. Change-Id: I5b228105ec719db89f0ab34b29fde11dbb1e5074 Reviewed-on: https://chrome-internal-review.googlesource.com/287195 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Yunlian Jiang <yunlian@google.com>
2016-09-14[crosperf] Don't run kernel test in llvm rotating testsers.Caroline Tice
Update the nightly rotating testers to not run the kernel tests for the LLVM images, since the kernel is not built with LLVM BUG=None TEST=None Change-Id: Ib2efe2211d6c9e86b80db2470c428d70e30e5516 Reviewed-on: https://chrome-internal-review.googlesource.com/286800 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Yunlian Jiang <yunlian@google.com>
2016-09-13crosperf+cros_utils: Swap to #!/usr/bin/env.George Burgess IV
`cros lint` was recently updated to complain about using #!/usr/bin/python{,2,3} directly. Instead, it prefers /usr/bin/env python{2,3} now. BUG=None TEST=./run_tests.sh passes; linter seems much happier with crosperf. Change-Id: I9b88a7af1f8e03b7c870781c3d8dc1b872a3dfc3 Reviewed-on: https://chrome-internal-review.googlesource.com/286341 Commit-Ready: Luis Lozano <llozano@chromium.org> Tested-by: Luis Lozano <llozano@chromium.org> Reviewed-by: Luis Lozano <llozano@chromium.org>
2016-09-12crosperf: Add generate_report.py.George Burgess IV
This script allows us to generate HTML, JSON, text, and email reports for JSON inputs, rather than needing to run things through crosperf. BUG=chromium:641098 TEST=cros lint on new files passes, ./run_tests.sh passes. Change-Id: I213149c0feca8b97f72fb447030c670e2f631a87 Reviewed-on: https://chrome-internal-review.googlesource.com/285238 Commit-Ready: George Burgess <gbiv@google.com> Tested-by: George Burgess <gbiv@google.com> Reviewed-by: Caroline Tice <cmtice@google.com>
2016-09-08Fix test failures in results_report refactors.George Burgess IV
...I have absolutely no clue how I missed these failures (since I had to run results_report_unittest to run the new perf parser test), but apparently I did. The failure happened because we had a _chart_js arg name, which didn't match up with chart_js (in the caller). This also fixes a few minor linter complaints that I introduced in patch-set 2 of said review. All of the other files I touched are clean (except for a complaint about use of _logger in one of them, but that was there before I made my changes). BUG=None TEST=./run_tests.sh *actually* passes this time. I promise. :) Change-Id: I98e43b29ff9cef5fc9cf33143adc26b98a861f3e Reviewed-on: https://chrome-internal-review.googlesource.com/285436 Commit-Ready: George Burgess <gbiv@google.com> Tested-by: George Burgess <gbiv@google.com> Reviewed-by: Caroline Tice <cmtice@google.com>
2016-09-08[Crosperf] Fix typo & update boards.Caroline Tice
This CL fixes a typo in the rotating tester tests, and updates the board list. Also reformats script (using tc_pyformat). BUG=None TEST=None Change-Id: I0a21bd8fd9d20d5d7dcd49e3484201c1860d352f Reviewed-on: https://chrome-internal-review.googlesource.com/285456 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Yunlian Jiang <yunlian@google.com>
2016-09-08crosperf: Make results_report more general.George Burgess IV
The goal of this patch is to allow us to use results_report to generate non-ChromeOS-specific results reports. It's meant to be a nop for regular crosperf runs (except that, in HTML report generation, whitespace in the HTML itself will look a bit different.) Moreover, results_report used to shuffle Experiments around, each of which, unsurprisingly, contained *tons* of data about an experiment. So, part of this patch was reducing results_report's reliance on Experiments, in favor of a new type, BenchmarkResult. Some parts of results_report still rely on Experiments, but only to provide *extra* data. The minimum amount of data needed to make a results report is contained in a BenchmarkResult, and there's a convenient API we can use to make a BenchmarkResult out of an Experiment. This patch also does a massive refactor of results_report, because lots of the code was mildly more icky than it is with this patch applied. The refactor-for-prettiness and refactor-for-BenchmarkResults kind of go hand-in-hand, so it's really hard to split them out. The only part that's not so difficult to split out is the refactor to results_report_templates, but the value of splitting that out is questionable (to me). Speaking of which, all HTML magicks that were in results_report is now in results_report_templates, and now use *actual* Templates. It -- and HTMLReportGenerator -- are hopefully more readable as a result. Next, this makes JSONRsultsReport's GetReport() return a string containing JSON data, rather than calling a callback to write that string to a file. Finally, this includes a change to perf_table. Namely, its removal. It was otherwise unused, and I couldn't get it to even work, so I cleaned it up, made a perf report parser (because I couldn't get the cros_utils one to work, either...), and made sure it functions. If we're able to produce perf reports that cros_utils can parse, I'm happy to back my new parser out; I was primarily using it so I could be sure I didn't break HTML report generation. BUG=chromium:641098 TEST=./run_tests.sh passes Change-Id: I437de9eb39e00c9dd5c223ecd27feaaab544a6fd Reviewed-on: https://chrome-internal-review.googlesource.com/282217 Commit-Ready: George Burgess <gbiv@google.com> Tested-by: George Burgess <gbiv@google.com> Reviewed-by: Caroline Tice <cmtice@google.com>
2016-09-02cros_utils: increast timeout for trybot.Yunlian Jiang
This increase time out for trybot jobs to 8 hours. BUG=None TEST=None Change-Id: Ieed04fa7082e4d71ee1ed9cee62e6eb33e6763a9 Reviewed-on: https://chrome-internal-review.googlesource.com/283936 Commit-Ready: Yunlian Jiang <yunlian@google.com> Tested-by: Yunlian Jiang <yunlian@google.com> Reviewed-by: Ting-Yuan Huang <laszio@google.com>
2016-08-31crosperf: fix appears to have succeeded but returned no results errorYunlian Jiang
This fixes the error appears to have succeeded but returned no results for a set of benchmarks. BUG=chromium:642525 TEST=the results appear for these benchmarks Change-Id: Icad1e1da5881f664cdd96b4cfc3e9ac590db087d Reviewed-on: https://chrome-internal-review.googlesource.com/282696 Commit-Ready: Yunlian Jiang <yunlian@google.com> Tested-by: Yunlian Jiang <yunlian@google.com> Reviewed-by: Caroline Tice <cmtice@google.com>
2016-08-29Update owner file to include everyone in the team.Luis Lozano
BUG=None TEST=None Change-Id: I1caa63c014921a09358cb115d56a3333fa858be9 Reviewed-on: https://chrome-internal-review.googlesource.com/281539 Commit-Ready: Luis Lozano <llozano@chromium.org> Tested-by: Luis Lozano <llozano@chromium.org> Reviewed-by: Caroline Tice <cmtice@google.com>
2016-08-29crosperf: Fix a failing test.George Burgess IV
An earlier commit added extra spacing between remotes in crosperf/default_remotes. This resulted in experiment_factory_unittests.py reporting that there were 5 remotes in some cases (2 of which were ''), since we were splitting on one space. It seems sensible to split on any amount of whitespace, instead. BUG=None TEST=./run_tests.sh passes. Change-Id: I95bd85ba92e5e1f5b30b0273b1b4a6c8853b1995 Reviewed-on: https://chrome-internal-review.googlesource.com/281538 Commit-Ready: George Burgess <gbiv@google.com> Tested-by: George Burgess <gbiv@google.com> Reviewed-by: Caroline Tice <cmtice@google.com>
2016-08-26crosperf: refactor results_organizer; NFC.George Burgess IV
As part of playing with results_report (so I could make it less ChromeOS-dependent), I started refactoring results_organizer, which is only used by results_report. This was the result. The goal was to make it easier to understand what, exactly, results_organizer is trying to do. BUG=chromium:641098 TEST=./test_all.sh passes. Change-Id: I8e53175fbc1ded6f23631550223905b3321376ca Reviewed-on: https://chrome-internal-review.googlesource.com/280815 Commit-Ready: George Burgess <gbiv@google.com> Tested-by: George Burgess <gbiv@google.com> Reviewed-by: Caroline Tice <cmtice@google.com>
2016-08-26crosperf: update lab machine names.Rahul Chaudhry
Updated default_remotes to reflect the latest locations of all toolchain machines from https://sites.google.com/a/google.com/englabs/resources/infrastructure/chromeos/chromeos-devices/special-duts/toolchain BUG=chromium:639389 TEST=None Change-Id: I1da60a961560551a6a63f92f38798f7153197089 Reviewed-on: https://chrome-internal-review.googlesource.com/281375 Commit-Ready: Rahul Chaudhry <rahulchaudhry@google.com> Tested-by: Rahul Chaudhry <rahulchaudhry@google.com> Reviewed-by: Caroline Tice <cmtice@google.com>
2016-08-24crosperf: fix telemetry results parsing.Yunlian Jiang
New telemetry results provide multiple values for a single key. This CL uses the average value to make crosperf work. BUG=chromium:639896 TEST=run crosperf with 2 iterations of speedometer. Change-Id: I32e7a00a57dbbe995915af0c32918f1d680ae484 Reviewed-on: https://chrome-internal-review.googlesource.com/279057 Commit-Ready: Yunlian Jiang <yunlian@google.com> Tested-by: Yunlian Jiang <yunlian@google.com> Reviewed-by: Caroline Tice <cmtice@google.com>
2016-08-19[crosperf] Fix reading from cache.Caroline Tice
The read-from-cache code needed to be updated to deal with the new telemetry results format, as did the code for generating json reports (for archiving in the database). BUG=chromium:639106 TEST=Tested with crosperf reading from cache & generating json results. Change-Id: I2c97e5b393583048cc74559d065f498c0f0675a1 Reviewed-on: https://chrome-internal-review.googlesource.com/278298 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Yunlian Jiang <yunlian@google.com>
2016-08-19[crosperf] Update lab machine names/locations.Caroline Tice
This CL updates the lab machine names & locations now that they have been moved. BUG=chromium:639389 TEST=pinged all the machines Change-Id: Id30c8635ea3b33e789294f731f28ee2a88b5f17a Reviewed-on: https://chrome-internal-review.googlesource.com/278119 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Yunlian Jiang <yunlian@google.com>
2016-08-18binary search tool: Update ChromeOS Object bisectorCassidy Burden
Update ChromeOS Object bisector to work with new compiler wrapper. This new compiler wrapper means we can now do incremental switches and triage builds that skip compiler calls. TEST=Run full bisection CQ-DEPEND=CL:368231 Change-Id: If34877519ad6dfe3189a5e934e174fd13dae5c89 Reviewed-on: https://chrome-internal-review.googlesource.com/274236 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Caroline Tice <cmtice@google.com>
2016-08-17[crosperf] Fix telemetry results parsing.Caroline Tice
Many telemetry results fields now contain multiple values. This CL fixes crosperf to accomodate that. BUG=None TEST=Ran crosperf with change and it worked. Change-Id: Iccb8926223c8447b72d51f413756f5dbc09f89de Reviewed-on: https://chrome-internal-review.googlesource.com/276817 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Luis Lozano <llozano@chromium.org>
2016-08-16[crosperf] Change 'page_cycler.' to 'page_cycler_v2.' everywhere.Caroline Tice
Telemetry is switching to using page_cyler_v2 and will stop supporting old page_cycler soon. This CL switches everything in toolchain-utils to reference page_cycler_v2. BUG=None TEST=Ran unittests in crosperf with no issues. Change-Id: I0e3ce29bfaf483507a9ba528c6c7c5a8fa00f00e Reviewed-on: https://chrome-internal-review.googlesource.com/271598 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Caroline Tice <cmtice@google.com>
2016-08-16binary search tool: Run tc_pyformat on all python scriptsCassidy Burden
REFACTOR WARNING! Run tc_pyformat on all python scripts. TEST=Run unit tests, Android bisection Change-Id: I9624e7be2670bdb29bcdb59404bf9e8c2f7d81c5 Reviewed-on: https://chrome-internal-review.googlesource.com/274615 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Caroline Tice <cmtice@google.com>
2016-08-12binary search tool: Add sample NDK bisectionCassidy Burden
Add sample NDK bisection that can be given to those inside or outside of Google. This sample includes a compressed source tree of the Teapot sample NDK app, and a script showing the full setup and process for running the bisection tool. It also includes a couple patches that show the necessary changes to the build system, and the "compiler error" that we will be bisecting. TEST=Run DO_BISECTION.sh Change-Id: I11279fe33e833b0dfe2fb147fde6f46b347d62be Reviewed-on: https://chrome-internal-review.googlesource.com/274145 Commit-Ready: Cassidy Burden <cburden@google.com> Tested-by: Cassidy Burden <cburden@google.com> Reviewed-by: Luis Lozano <llozano@chromium.org>
2016-08-11binary search tool: Make 1st resumed iteration non-incrementalCassidy Burden
If resuming a bisection, we want to make sure all current good items are switched to good and all current bad items are switched to bad. This is so the bisection tool will resume in a known, stable state. Incremental builds are essentially diffs from previous builds, and the base environment may have changed since the last execution. TEST=Add test code, run unit tests Change-Id: I9ac855efbcfb48bc86f245f33ee29d0c38aa7af9 Reviewed-on: https://chrome-internal-review.googlesource.com/274175 Commit-Ready: Cassidy Burden <cburden@google.com> Tested-by: Cassidy Burden <cburden@google.com> Reviewed-by: Caroline Tice <cmtice@google.com>
2016-08-10[crosperf] Update parsing of telemetry json file.Caroline Tice
They've updated the format of the output in the json file. This updates crosperf to work with either the old or the new format. BUG=None TEST=./run_tests and ran crosperf successfully. Change-Id: I0b22f4c740943c55c0c521ab715e0b70fbbdfc69 Reviewed-on: https://chrome-internal-review.googlesource.com/274195 Commit-Ready: Caroline Tice <cmtice@google.com> Tested-by: Caroline Tice <cmtice@google.com> Reviewed-by: Yunlian Jiang <yunlian@google.com>
2016-08-10Random nit fixes in crosperf; NFC.George Burgess IV
This also removes results_sorter.py, which seems to be unused and untested. TEST=./run_tests.sh passes BUG=None Change-Id: I926affb23c78a8a0aedea4ab279da590f8cf93b5 Reviewed-on: https://chrome-internal-review.googlesource.com/273995 Commit-Ready: George Burgess <gbiv@google.com> Tested-by: George Burgess <gbiv@google.com> Reviewed-by: Caroline Tice <cmtice@google.com>
2016-08-10binary search tool: Add maintenance documentCassidy Burden
Change-Id: I2c7e10f4f0622569490f97fd9406e1deb5940041 Reviewed-on: https://chrome-internal-review.googlesource.com/273451 Commit-Ready: Cassidy Burden <cburden@google.com> Tested-by: Cassidy Burden <cburden@google.com> Reviewed-by: Caroline Tice <cmtice@google.com>
2016-08-09binary search tool: Fix edge cases in compiler wrapperCassidy Burden
Fix compiler wrapper edge cases that exist with the NDK and ChromeOS build systems. For ChromeOS we support the -MMD flag when caching dependency files. For NDK we support calls to compiler that give no output file (these calls are meant to check compiler capabilities). TEST=NDK bisection and ChromeOS population Change-Id: Ic23bc7876e9844eacb106d49082e0509ea1ee31b Reviewed-on: https://chrome-internal-review.googlesource.com/273985 Commit-Ready: Cassidy Burden <cburden@google.com> Tested-by: Cassidy Burden <cburden@google.com> Reviewed-by: Caroline Tice <cmtice@google.com>
generated by cgit v1.2.3 (git 2.25.1) at 2024-07-11 09:38:12 +0000